DanielLeens commented on issue #11769: URL: https://github.com/apache/seatunnel/issues/11769#issuecomment-5425086877
Thanks for confirming that the issue still reproduces on tag `2.3.12` even with #9942 and #10793 applied manually. That is useful because it shows those two patches are still not sufficient for your scenario. At the same time, it is still different from validating the current `dev` branch or the active fix branch directly, so we should keep the merge decision tied to the current repair path rather than to a 2.3.12 cherry-pick result alone. For this thread the active path is still: 1. #11879 for the regression test covering the multi-table replay path; 2. #11883 as the current functional-fix candidate for the stale-statement / connection-leak behavior. So yes, #11883 remains the right PR to evaluate here. If you can, one more very helpful check would be: 1. the exact failing log snippet around the `08S01` / reconnect / replay step from your test run; 2. whether the same scenario still reproduces when you validate against the current #11883 branch, or a `dev` build that includes it. That will let us compare the remaining runtime gap directly against the active fix candidate before merge. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
